CVPR 2018 论文列表

We address the computational problem of novel human pose synthesis. Given an image of a person and a desired pose, we produce a depiction of that person in that pose, retaining the appearance of both the person and background. We present a modular generative neural network that synthesizes unseen poses using training pairs of images and poses taken from human action videos. Our network separates a scene into different body part and background layers, moves body parts to new locations and refines their appearances, and composites the new foreground with a hole-filled background. These subtasks, implemented with separate modules, are trained jointly using only a single target image as a supervised label. We use an adversarial discriminator to force our network to synthesize realistic details conditioned on pose. We demonstrate image synthesis results on three action classes: golf, yoga/workouts and tennis, and show that our method produces accurate results within action classes as well as across action classes. Given a sequence of desired poses, we also produce coherent videos of actions.

We study data-driven representations for three-dimensional triangle meshes, which are one of the prevalent objects used to represent 3D geometry. Recent works have developed models that exploit the intrinsic geometry of manifolds and graphs, namely the Graph Neural Networks (GNNs) and its spectral variants, which learn from the local metric tensor via the Laplacian operator. Despite offering excellent sample complexity and built-in invariances, intrinsic geometry alone is invariant to isometric deformations, making it unsuitable for many applications. To overcome this limitation, we propose several upgrades to GNNs to leverage extrinsic differential geometry properties of three-dimensional surfaces, increasing its modeling power. In particular, we propose to exploit the Dirac operator, whose spectrum detects principal curvature directions --- this is in stark contrast with the classical Laplace operator, which directly measures mean curvature. We coin the resulting models emph{Surface Networks (SN)}. We prove that these models define shape representations that are stable to deformation and to discretization, and we demonstrate the efficiency and versatility of SNs on two challenging tasks: temporal prediction of mesh deformations under non-linear dynamics and generative models using a variational autoencoder framework with encoders/decoders given by SNs.

The last few years have seen approaches trying to combine the increasing popularity of depth sensors and the success of the convolutional neural networks. Using depth as additional channel alongside the RGB input has the scale variance problem present in image convolution based approaches. On the other hand, 3D convolution wastes a large amount of memory on mostly unoccupied 3D space, which consists of only the surface visible to the sensor. Instead, we propose SurfConv, which “slides” compact 2D filters along the visible 3D surface. SurfConv is formulated as a simple depth-aware multi-scale 2D convolution, through a new Data-Driven Depth Discretization (D4) scheme. We demonstrate the effectiveness of our method on indoor and outdoor 3D semantic segmentation datasets. Our method achieves state-of-the-art performance while using less than 30% parameters used by the 3D convolution based approaches.

In this paper, we present supervision-by-registration, an unsupervised approach to improve the precision of facial landmark detectors on both images and video. Our key observation is that the detections of the same landmark in adjacent frames should be coherent with registration, i.e., optical flow. Interestingly, coherency of optical flow is a source of supervision that does not require manual labeling, and can be leveraged during detector training. For example, we can enforce in the training loss function that a detected landmark at frame t-1 followed by optical flow tracking from frame t-1 to frame t should coincide with the location of the detection at frame t. Essentially, supervision-by-registration augments the training loss function with a registration loss, thus training the detector to have output that is not only close to the annotations in labeled images, but also consistent with registration on large amounts of unlabeled videos. End-to-end training with the registration loss is made possible by a differentiable Lucas-Kanade operation, which computes optical flow registration in the forward pass, and back-propagates gradients that encourage temporal coherency in the detector. The output of our method is a more precise image-based facial landmark detector, which can be applied to single images or video. With supervision-by-registration, we demonstrate (1) improvements in facial landmark detection on both images (300W, ALFW) and video (300VW, Youtube-Celebrities), and (2) significant reduction of jittering in video detections.

Given a tiny face image, conventional face hallucination methods aim to super-resolve its high-resolution (HR) counterpart by learning a mapping from an exemplar dataset. Since a low-resolution (LR) input patch may correspond to many HR candidate patches, this ambiguity may lead to erroneous HR facial details and thus distorts final results, such as gender reversal. An LR input contains low-frequency facial components of its HR version while its residual face image defined as the difference between the HR ground-truth and interpolated LR images contains the missing high-frequency facial details. We demonstrate that supplementing residual images or feature maps with facial attribute information can significantly reduce the ambiguity in face super-resolution. To explore this idea, we develop an attribute-embedded upsampling network, which consists of an upsampling network and a discriminative network. The upsampling network is composed of an autoencoder with skip-connections, which incorporates facial attribute vectors into the residual features of LR inputs at the bottleneck of the autoencoder and deconvolutional layers used for upsampling. The discriminative network is designed to examine whether super-resolved faces contain the desired attributes or not and then its loss is used for updating the upsampling network. In this manner, we can super-resolve tiny unaligned (16$ imes$16 pixels) face images with a large upscaling factor of 8$ imes$ while reducing the uncertainty of one-to-many mappings significantly. By conducting extensive evaluations on a large-scale dataset, we demonstrate that our method achieves superior face hallucination results and outperforms the state-of-the-art.

This paper addresses 2 challenging tasks: improving the quality of low resolution facial images and accurately locating the facial landmarks on such poor resolution images. To this end, we make the following 5 contributions: (a) we propose Super-FAN: the very first end-to-end system that addresses both tasks simultaneously, i.e. both improves face resolution and detects the facial landmarks. The novelty or Super-FAN lies in incorporating structural information in a GAN-based super-resolution algorithm via integrating a sub-network for face alignment through heatmap regression and optimizing a novel heatmap loss. (b) We illustrate the benefit of training the two networks jointly by reporting good results not only on frontal images (as in prior work) but on the whole spectrum of facial poses, and not only on synthetic low resolution images (as in prior work) but also on real-world images. (c) We improve upon the state-of-the-art in face super-resolution by proposing a new residual-based architecture. (d) Quantitatively, we show large improvement over the state-of-the-art for both face super-resolution and alignment. (e) Qualitatively, we show for the first time good results on real-world low resolution images.

Given two consecutive frames, video interpolation aims at generating intermediate frame(s) to form both spatially and temporally coherent video sequences. While most existing methods focus on single-frame interpolation, we propose an end-to-end convolutional neural network for variable-length multi-frame video interpolation, where the motion interpretation and occlusion reasoning are jointly modeled. We start by computing bi-directional optical flow between the input images using a U-Net architecture. These flows are then linearly combined at each time step to approximate the intermediate bi-directional optical flows. These approximate flows, however, only work well in locally smooth regions and produce artifacts around motion boundaries. To address this shortcoming, we employ another U-Net to refine the approximated flow and also predict soft visibility maps. Finally, the two input images are warped and linearly fused to form each intermediate frame. By applying the visibility maps to the warped images before fusion, we exclude the contribution of occluded pixels to the interpolated intermediate frame to avoid artifacts. Since none of our learned network parameters are time-dependent, our approach is able to produce as many intermediate frames as needed. To train our network, we use 1,132 240-fps video clips, containing 300K individual video frames. Experimental results on several datasets, predicting different numbers of interpolated frames, demonstrate that our approach performs consistently better than existing methods.

Recent advances in facial landmark detection achieve success by learning discriminative features from rich deformation of face shapes and poses. Besides the variance of faces themselves, the intrinsic variance of image styles, e.g., grayscale vs. color images, light vs. dark, intense vs. dull, and so on, has constantly been overlooked. This issue becomes inevitable as increasing web images are collected from various sources for training neural networks. In this work, we propose a style-aggregated approach to deal with the large intrinsic variance of image styles for facial landmark detection. Our method transforms original face images to style-aggregated images by a generative adversarial module. The proposed scheme uses the style-aggregated image to maintain face images that are more robust to environmental changes. Then the original face images accompanying with style-aggregated ones play a duet to train a landmark detector which is complementary to each other. In this way, for each face, our method takes two images as input, i.e., one in its original style and the other in the aggregated style. In experiments, we observe that the large variance of image styles would degenerate the performance of facial landmark detectors. Moreover, we show the robustness of our method to the large variance of image styles by comparing to a variant of our approach, in which the generative adversarial module is removed, and no style-aggregated images are used. Our approach is demonstrated to perform well when compared with state-of-the-art algorithms on benchmark datasets AFLW and 300-W. Code is publicly available on GitHub: https://github.com/D-X-Y/SAN

This paper is the first work to propose a network to predict a structured uncertainty distribution for a synthesized image. Previous approaches have been mostly limited to predicting diagonal covariance matrices. Our novel model learns to predict a full Gaussian covariance matrix for each reconstruction, which permits efficient sampling and likelihood evaluation. We demonstrate that our model can accurately reconstruct ground truth correlated residual distributions for synthetic datasets and generate plausible high frequency samples for real face images. We also illustrate the use of these predicted covariances for structure preserving image denoising.

Diagrams often depict complex phenomena and serve as a good test bed for visual and textual reasoning. However, understanding diagrams using natural image understanding approaches requires large training datasets of diagrams, which are very hard to obtain. Instead, this can be addressed as a matching problem either between labeled diagrams, images or both. This problem is very challenging since the absence of significant color and texture renders local cues ambiguous and requires global reasoning. We consider the problem of one-shot part labeling: labeling multiple parts of an object in a target image given only a single source image of that category. For this set-to-set matching problem, we introduce the Structured Set Matching Network (SSMN), a structured prediction model that incorporates convolutional neural networks. The SSMN is trained using global normalization to maximize local match scores between corresponding elements and a global consistency score among all matched elements, while also enforcing a matching constraint between the two sets. The SSMN significantly outperforms several strong baselines on three label transfer scenarios: diagram-to-diagram, evaluated on a new diagram dataset of over 200 categories; image-to-image, evaluated on a dataset built on top of the Pascal Part Dataset; and image-to-diagram, evaluated on transferring labels across these datasets.

Recent works have shown the benefit of integrating Conditional Random Fields (CRFs) models into deep architectures for improving pixel-level prediction tasks. Following this line of research, in this paper we introduce a novel approach for monocular depth estimation. Similarly to previous works, our method employs a continuous CRF to fuse multi-scale information derived from different layers of a front-end Convolutional Neural Network (CNN). Differently from past works, our approach benefits from a structured attention model which automatically regulates the amount of information transferred between corresponding features at different scales. Importantly, the proposed attention model is seamlessly integrated into the CRF, allowing end-to-end training of the entire architecture. Our extensive experimental evaluation demonstrates the effectiveness of the proposed method which is competitive with previous methods on the KITTI benchmark and outperforms the state of the art on the NYU Depth V2 dataset.

Despite recent emergence of adversarial based methods for video prediction, existing algorithms often produce unsatisfied results in image regions with rich structural information (i.e., object boundary) and detailed motion (i.e., articulated body movement). To this end, we present a structure preserving video prediction framework to explicitly address above issues and enhance video prediction quality. On one hand, our framework contains a two-stream generation architecture which deals with high frequency video content (i.e., detailed object or articulated motion structure) and low frequency video content (i.e., location or moving directions) in two separate streams. On the other hand, we propose a RNN structure for video prediction, which employs temporal-adaptive convolutional kernels to capture time-varying motion patterns as well as the tiny object within a scene. Extensive experiments on diverse scene, ranging from human motion to semantic layout prediction, demonstrate the effectiveness of the proposed video prediction approach.

Context is important for accurate visual recognition. In this work we propose an object detection algorithm that not only considers object visual appearance, but also makes use of two kinds of context including scene contextual information and object relationships within a single image. Therefore, object detection is regarded as both a cognition problem and a reasoning problem when leveraging these structured information. Specifically, this paper formulates object detection as a problem of graph structure inference, where given an image the objects are treated as nodes in a graph and relationships between the objects are modeled as edges in such graph. To this end, we present a so-called Structure Inference Network (SIN), a detector that incorporates into a typical detection framework (e.g. Faster R-CNN) with a graphical model which aims to infer object state. Comprehensive experiments on PASCAL VOC and MS COCO datasets indicate that scene context and object relationships truly improve the performance of object detection with more desirable and reasonable outputs.

This paper proposes a new method for Non-rigidstructure-from-motion (NRSfM). Departing significantlyfrom the traditional idea of using linear low-order shapemodel for NRSfM, our method exploits the property of shaperecurrence (i.e. many dynamic shapes tend to repeat them-selves in time). We show that recurrency is in fact agen-eralized rigidity. Based on this, we show how to reduceNRSfM problems to rigid ones, provided that the recurrencecondition is satisfied. Given such a reduction, standardrigid-SFM techniques can be applied directly (without anychange) to reconstruct the non-rigid dynamic shape. To im-plement this idea as a practical approach, this paper de-velops efficient and reliable algorithm for automatic recur-rence detection, as well as new method for camera viewsclustering via rigidity-check. Experiments on both syntheticsequences and real data demonstrate the effectiveness of theproposed method. Since the method provides novel perspec-tive to look at Structure-from-Motion, we hope it will inspireother new researches in the field.

Variational inference has experienced a recent surge in popularity owing to stochastic approaches, which have yielded practical tools for a wide range of model classes. A key benefit is that stochastic variational inference obviates the tedious process of deriving analytical expressions for closed-form variable updates. Instead, one simply needs to derive the gradient of the log-posterior, which is often much easier. Yet for certain model classes, the log-posterior itself is difficult to optimize using standard gradient techniques. One such example are random field models, where optimization based on gradient linearization has proven popular, since it speeds up convergence significantly and can avoid poor local optima. In this paper we propose stochastic variational inference with gradient linearization (SVIGL). It is similarly convenient as standard stochastic variational inference - all that is required is a local linearization of the energy gradient. Its benefit over stochastic variational inference with conventional gradient methods is a clear improvement in convergence speed, while yielding comparable or even better variational approximations in terms of KL divergence. We demonstrate the benefits of SVIGL in three applications: Optical flow estimation, Poisson-Gaussian denoising, and 3D surface reconstruction.

It is desirable to train convolutional networks (CNNs) to run more efficiently during inference. In many cases however, the computational budget that the system has for inference cannot be known beforehand during training, or the inference budget is dependent on the changing real-time resource availability. Thus, it is inadequate to train just inference-efficient CNNs, whose inference costs are not adjustable and cannot adapt to varied inference budgets. We propose a novel approach for cost-adjustable inference in CNNs - Stochastic Downsampling Point (SDPoint). During training, SDPoint applies feature map downsampling to a random point in the layer hierarchy, with a random downsampling ratio. The different stochastic downsampling configurations known as SDPoint instances (of the same model) have computational costs different from each other, while being trained to minimize the same prediction loss. Sharing network parameters across different instances provides significant regularization boost. During inference, one may handpick a SDPoint instance that best fits the inference budget. The effectiveness of SDPoint, as both a cost-adjustable inference approach and a regularizer, is validated through extensive experiments on image classification.

This paper presents the first attempt at stereoscopic neural style transfer, which responds to the emerging demand for 3D movies or AR/VR. We start with a careful examination of applying existing monocular style transfer methods to left and right views of stereoscopic images separately. This reveals that the original disparity consistency cannot be well preserved in the final stylization results, which causes 3D fatigue to the viewers. To address this issue, we incorporate a new disparity loss into the widely adopted style loss function by enforcing the bidirectional disparity constraint in non-occluded regions. For a practical real-time solution, we propose the first feed-forward network by jointly training a stylization sub-network and a disparity sub-network, and integrate them in a feature level middle domain. Our disparity sub-network is also the first end-to-end network for simultaneous bidirectional disparity and occlusion mask estimation. Finally, our network is effectively extended to stereoscopic videos, by considering both temporal coherence and disparity consistency. We will show that the proposed method clearly outperforms the baseline algorithms both quantitatively and qualitatively.

We achieve tomography of 3D volumetric natural objects, where each projected 2D image corresponds to a different specimen. Each specimen has unknown random 3D orientation, location, and scale. This imaging scenario is relevant to microscopic and mesoscopic organisms, aerosols and hydrosols viewed naturally by a microscope. In-class scale variation inhibits prior single-particle reconstruction methods. We thus generalize tomographic recovery to account for all degrees of freedom of a similarity transformation. This enables geometric self-calibration in imaging of transparent objects. We make the computational load manageable and reach good quality reconstruction in a short time. This enables extraction of statistics that are important for a scientific study of specimen populations, specifically size distribution parameters. We apply the method to study of plankton.

Recent studies have shown remarkable success in image-to-image translation for two domains. However, existing approaches have limited scalability and robustness in handling more than two domains, since different models should be built independently for every pair of image domains. To address this limitation, we propose StarGAN, a novel and scalable approach that can perform image-to-image translations for multiple domains using only a single model. Such a unified model architecture of StarGAN allows simultaneous training of multiple datasets with different domains within a single network. This leads to StarGAN's superior quality of translated images compared to existing models as well as the novel capability of flexibly translating an input image to any desired target domain. We empirically demonstrate the effectiveness of our approach on a facial attribute transfer and a facial expression synthesis tasks.